On the first day of the critical reasoning classes that I teach for college freshmen, I tell students, with only a little hyperbole, that the less I talk, the more they will learn. They will not only be talking to me but to each other. The best way to become better at reasoning is to practice reasoning, and the best way to practice reasoning is to talk with others, to share one’s reasons and respond to the reasons of others.

My opening message is welcome insofar as it promises a less monotonous classroom. But, as students soon discover, it is also daunting. Reasoning together is hard work. And K-12 schooling in the United States has done little to prepare most us for it. In many respects, it has habituated us against it.

The “recitation” model of classroom talk

Consider the following transcription of a recorded classroom interaction that was studied by Ian A. G. Wilkinson and Alina Reznitskaya as part of the 2012-2015 research program “Dialogic Teaching: Professional Development in Classroom Discussion to Improve Students’ Argument Literacy.” In this exchange, presented in Reznitskaya and Wilkinson’s The Most Reasonable Answer (2017), a teacher talks with students (whose names have been replaced with pseudonyms) about a Native American narrative called “Tonweya and the Eagles” in which a young man’s quest lands him in a precarious situation:

Teacher: Okay. So, now he got stuck on the cliff. So, now what? That’s a big problem. . . . But, first of all, why did he want to reach the eagles? Gabriel?

Gabriel: Because he wanted to bring them back to his tribe so that everyone would have, like, a feather for everybody.

Teacher: Okay. He didn’t want to bring the eagles back. He wanted to bring what back, Trisha?

Trisha: The feathers.

Teacher: The feathers. For what? What’s it called? For what headgear? Who’s that person? What are they called? Andrew?

Andrew: The chief.

Teacher: The chief. The starts with a w?

Jack: Warriors.

Teacher. Warriors. For the warriors’ headgear. . . .

Anyone who has ever anxiously avoided a teacher’s eyes while desperately hoping that a neighboring classmate gets “called on” can appreciate how this kind of classroom environment discourages students from taking intellectual risks. If you respond with the wrong answer, you face embarrassment in front of your peers and an authority figure. Whereas if you remain silent, a more informed or self-assured peer--or, eventually, an exasperated teacher--will disclose the information you’ll need for that test later.

The pattern exemplified by the exchange above, in which a teacher initiates a question, students respond, and the teacher evaluates and provides feedback on that response, has been called IRE or IRF--for initiation, response, and evaluation or initiation, response, and feedback--or simply recitation. Generations of observers of American education have found that recitation is the “default option,” as the applied linguist Courtney Cazden put it in her influential book Classroom Discourse (2001, p. 31). Many teachers who report that they use “discussion” as a teaching technique appear when observed to be relying on recitation (Alvermann, O’Brien, & Dillon, 1990).

Martin Nystrand and colleagues (1997) observed hundreds of 8th and 9th grade English classes over two years and coded more than 20,000 questions for their source (whether initiated by a student or teacher), response (whether the question was answered or not), authenticity (whether an answer was prespecified or known in advance by the teacher), uptake(the “incorporation of a previous answer into a subsequent question”), cognitive level (the “type of cognitive demand made by the question”), and level of evaluation (whether the teacher valorized and elaborated the students responses”) (pp. 37-38).

The study found that 85% of instructional time was devoted to some combination of lecture, recitation, and seatwork (p. 42). Student-initiated discussion of authentic questions was almost nonexistent: on average, "discussion took 50 seconds per class in eighth grade and less than 15 seconds in grade nine; small-group work, which occupied about half a minute in eighth grade, took a bit more than 2 minutes in 9th grade" (Nystrand 1997, p. 42). Nystrand notes that these findings corroborate numerous studies since the 1960s that have found American classroom talk to be dominated by teacher-driven questions that “test students’ recall of textbook information in recitation format” (Nystrand, 2006, p. 395).

The lessons of recitation

The reliance on recitation is understandable. As Nystrand notes, commentators in the early 20th century remarked on the difference between American classrooms, with their use of recitation, and European classrooms, where lecture remained the primary mode of instruction. In this context, recitation could be regarded as a more inclusive and democratic alternative to tradition (p. 394). Given the importance of “talking to learn” (Britton, 1969), we can appreciate the fact that students in an IRE classroom are not entirely silent at least. Moreover, considering teachers' limited time and the institutional pressure to cover the material needed to prepare students for high-stakes tests, it is hardly surprising that most would neglect student-initiated discussion of authentic questions.

Nevertheless, we must reckon with what recitation teaches, both explicitly, by the pedagogical aims that it favors, and implicitly, by the norms of conversation and community that it models.

The first thing to notice about the questions found in the interaction above (Did Tonweya want to bring back eagles or just eagle feathers? For whom?) is that they are not particularly challenging. The cognitive task such questions are designed to get students to perform, recalling facts about a text, is located at what educators would call the lowest levels of Bloom’s taxonomy of types of learning objectives; namely, remembering and comprehending in contrast to the higher-order tasks of applying, analyzing, synthesizing, evaluating, or creating.

Furthermore, all of the questions are aimed at eliciting information that the teacher already possesses. Students are not being prompted, for example, to offer their interpretation of the text or evaluation of its quality as a piece of literary art, let alone to explain how it might connect to their own experience. No one is asked to share reasons for their views. Since the answers to the questions are predetermined by the asker, it does not matter who answers. Anyone with the correct information can supply them. The voice and perspective of any particular student has no distinctive value.

Relatedly, it makes no difference what anyone says. Whether or not a student could be prompted to say so, the teacher was going to make sure the class heard that the feathers were for warriors’ headgear. While they provide some evidence of at least some students’ grasp of a text, IRE interactions do not give each participant some power to shape the direction and outcome of the talk. Control belongs almost entirely with the teacher.

Finally, participants in this talk are not thinking together. Students do not talk to one another but instead only address the teacher. They do not ask questions. They offer no reasons for their views. And rather than building upon or challenging each other’s contributions, they engage in parallel but essentially individual--and potentially competitive--interactions with the instructor. While many would call this classroom talk a “discussion,” it is not a dialogue.

Recitation favors declarative knowledge over analysis, evaluation, and judgment. It encourages individual performance in place of public deliberation. And it rewards those participants who are already confident in the correct answer rather than empowering all participants to explore an open question through collaborative reasoning. Above all, it tells students they do not have a real say in their learning community. Is it any wonder that so many college freshmen don’t feel ready to talk?